From: Sandipan Das <sandipan.das@amd.com>
To: <linux-perf-users@vger.kernel.org>, <linux-kernel@vger.kernel.org>
Cc: <peterz@infradead.org>, <mingo@redhat.com>, <acme@kernel.org>,
<namhyung@kernel.org>, <mark.rutland@arm.com>,
<alexander.shishkin@linux.intel.com>, <jolsa@kernel.org>,
<irogers@google.com>, <adrian.hunter@intel.com>,
<kan.liang@linux.intel.com>, <tglx@linutronix.de>, <bp@alien8.de>,
<dave.hansen@linux.intel.com>, <x86@kernel.org>, <hpa@zytor.com>,
<eranian@google.com>, <songliubraving@meta.com>,
<ravi.bangoria@amd.com>, <ananth.narayan@amd.com>,
<sandipan.das@amd.com>
Subject: [PATCH 2/4] perf/x86/amd/uncore: Use hrtimer for handling overflows
Date: Wed, 9 Apr 2025 13:27:07 +0530 [thread overview]
Message-ID: <f3a7703c22e6734f0d1bf34bc56be3124f818a8b.1744184837.git.sandipan.das@amd.com> (raw)
In-Reply-To: <cover.1744184837.git.sandipan.das@amd.com>
Uncore counters do not provide mechanisms like interrupts to report
overflows and the accumulated user-visible count is incorrect if there
is more than one overflow between two successive read requests for the
same event because the value of prev_count goes out-of-date for
calculating the correct delta.
To avoid this, start a hrtimer to periodically initiate a pmu->read() of
the active counters for keeping prev_count up-to-date. It should be
noted that the hrtimer duration should be lesser than the shortest time
it takes for a counter to overflow for this approach to be effective.
Signed-off-by: Sandipan Das <sandipan.das@amd.com>
---
arch/x86/events/amd/uncore.c | 72 ++++++++++++++++++++++++++++++++++++
1 file changed, 72 insertions(+)
diff --git a/arch/x86/events/amd/uncore.c b/arch/x86/events/amd/uncore.c
index 010024f09f2c..ee6528f2189f 100644
--- a/arch/x86/events/amd/uncore.c
+++ b/arch/x86/events/amd/uncore.c
@@ -21,6 +21,7 @@
#define NUM_COUNTERS_NB 4
#define NUM_COUNTERS_L2 4
#define NUM_COUNTERS_L3 6
+#define NUM_COUNTERS_MAX 64
#define RDPMC_BASE_NB 6
#define RDPMC_BASE_LLC 10
@@ -38,6 +39,10 @@ struct amd_uncore_ctx {
int refcnt;
int cpu;
struct perf_event **events;
+ unsigned long active_mask[BITS_TO_LONGS(NUM_COUNTERS_MAX)];
+ int nr_active;
+ struct hrtimer hrtimer;
+ u64 hrtimer_duration;
};
struct amd_uncore_pmu {
@@ -87,6 +92,51 @@ static struct amd_uncore_pmu *event_to_amd_uncore_pmu(struct perf_event *event)
return container_of(event->pmu, struct amd_uncore_pmu, pmu);
}
+static enum hrtimer_restart amd_uncore_hrtimer(struct hrtimer *hrtimer)
+{
+ struct amd_uncore_ctx *ctx;
+ struct perf_event *event;
+ unsigned long flags;
+ int bit;
+
+ ctx = container_of(hrtimer, struct amd_uncore_ctx, hrtimer);
+
+ if (!ctx->nr_active || ctx->cpu != smp_processor_id())
+ return HRTIMER_NORESTART;
+
+ /*
+ * Disable local interrupts to prevent pmu->start() or pmu->stop()
+ * from interrupting the update process
+ */
+ local_irq_save(flags);
+
+ for_each_set_bit(bit, ctx->active_mask, NUM_COUNTERS_MAX) {
+ event = ctx->events[bit];
+ event->pmu->read(event);
+ }
+
+ local_irq_restore(flags);
+
+ hrtimer_forward_now(hrtimer, ns_to_ktime(ctx->hrtimer_duration));
+ return HRTIMER_RESTART;
+}
+
+static void amd_uncore_start_hrtimer(struct amd_uncore_ctx *ctx)
+{
+ hrtimer_start(&ctx->hrtimer, ns_to_ktime(ctx->hrtimer_duration),
+ HRTIMER_MODE_REL_PINNED);
+}
+
+static void amd_uncore_cancel_hrtimer(struct amd_uncore_ctx *ctx)
+{
+ hrtimer_cancel(&ctx->hrtimer);
+}
+
+static void amd_uncore_init_hrtimer(struct amd_uncore_ctx *ctx)
+{
+ hrtimer_setup(&ctx->hrtimer, amd_uncore_hrtimer, CLOCK_MONOTONIC, HRTIMER_MODE_REL);
+}
+
static void amd_uncore_read(struct perf_event *event)
{
struct hw_perf_event *hwc = &event->hw;
@@ -117,18 +167,26 @@ static void amd_uncore_read(struct perf_event *event)
static void amd_uncore_start(struct perf_event *event, int flags)
{
+ struct amd_uncore_pmu *pmu = event_to_amd_uncore_pmu(event);
+ struct amd_uncore_ctx *ctx = *per_cpu_ptr(pmu->ctx, event->cpu);
struct hw_perf_event *hwc = &event->hw;
+ if (!ctx->nr_active++)
+ amd_uncore_start_hrtimer(ctx);
+
if (flags & PERF_EF_RELOAD)
wrmsrl(hwc->event_base, (u64)local64_read(&hwc->prev_count));
hwc->state = 0;
+ __set_bit(hwc->idx, ctx->active_mask);
wrmsrl(hwc->config_base, (hwc->config | ARCH_PERFMON_EVENTSEL_ENABLE));
perf_event_update_userpage(event);
}
static void amd_uncore_stop(struct perf_event *event, int flags)
{
+ struct amd_uncore_pmu *pmu = event_to_amd_uncore_pmu(event);
+ struct amd_uncore_ctx *ctx = *per_cpu_ptr(pmu->ctx, event->cpu);
struct hw_perf_event *hwc = &event->hw;
wrmsrl(hwc->config_base, hwc->config);
@@ -138,6 +196,11 @@ static void amd_uncore_stop(struct perf_event *event, int flags)
event->pmu->read(event);
hwc->state |= PERF_HES_UPTODATE;
}
+
+ if (!--ctx->nr_active)
+ amd_uncore_cancel_hrtimer(ctx);
+
+ __clear_bit(hwc->idx, ctx->active_mask);
}
static int amd_uncore_add(struct perf_event *event, int flags)
@@ -490,6 +553,9 @@ static int amd_uncore_ctx_init(struct amd_uncore *uncore, unsigned int cpu)
goto fail;
}
+ amd_uncore_init_hrtimer(curr);
+ curr->hrtimer_duration = 60LL * NSEC_PER_SEC;
+
cpumask_set_cpu(cpu, &pmu->active_mask);
}
@@ -879,12 +945,18 @@ static int amd_uncore_umc_event_init(struct perf_event *event)
static void amd_uncore_umc_start(struct perf_event *event, int flags)
{
+ struct amd_uncore_pmu *pmu = event_to_amd_uncore_pmu(event);
+ struct amd_uncore_ctx *ctx = *per_cpu_ptr(pmu->ctx, event->cpu);
struct hw_perf_event *hwc = &event->hw;
+ if (!ctx->nr_active++)
+ amd_uncore_start_hrtimer(ctx);
+
if (flags & PERF_EF_RELOAD)
wrmsrl(hwc->event_base, (u64)local64_read(&hwc->prev_count));
hwc->state = 0;
+ __set_bit(hwc->idx, ctx->active_mask);
wrmsrl(hwc->config_base, (hwc->config | AMD64_PERFMON_V2_ENABLE_UMC));
perf_event_update_userpage(event);
}
--
2.43.0
next prev parent reply other threads:[~2025-04-09 7:58 UTC|newest]
Thread overview: 8+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-04-09 7:57 [PATCH 0/4] perf/x86/amd/uncore: Overflow handling enhancements Sandipan Das
2025-04-09 7:57 ` [PATCH 1/4] perf/x86/amd/uncore: Remove unused member from amd_uncore_ctx Sandipan Das
2025-04-09 7:57 ` Sandipan Das [this message]
2025-04-10 8:20 ` [PATCH 2/4] perf/x86/amd/uncore: Use hrtimer for handling overflows Peter Zijlstra
2025-04-10 8:40 ` Sandipan Das
2025-04-09 7:57 ` [PATCH 3/4] perf/x86/amd/uncore: Add parameter to configure hrtimer Sandipan Das
2025-04-09 7:57 ` [PATCH 4/4] perf/x86/amd/uncore: Prevent UMC counters from saturating Sandipan Das
2025-04-16 19:15 ` Song Liu
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=f3a7703c22e6734f0d1bf34bc56be3124f818a8b.1744184837.git.sandipan.das@amd.com \
--to=sandipan.das@amd.com \
--cc=acme@kernel.org \
--cc=adrian.hunter@intel.com \
--cc=alexander.shishkin@linux.intel.com \
--cc=ananth.narayan@amd.com \
--cc=bp@alien8.de \
--cc=dave.hansen@linux.intel.com \
--cc=eranian@google.com \
--cc=hpa@zytor.com \
--cc=irogers@google.com \
--cc=jolsa@kernel.org \
--cc=kan.liang@linux.intel.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-perf-users@vger.kernel.org \
--cc=mark.rutland@arm.com \
--cc=mingo@redhat.com \
--cc=namhyung@kernel.org \
--cc=peterz@infradead.org \
--cc=ravi.bangoria@amd.com \
--cc=songliubraving@meta.com \
--cc=tglx@linutronix.de \
--cc=x86@kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox
all inboxes | Powered by JetHome®